1

I am a little confused by a suggestion for using Time on Task as a metric for user testing. In the past I have only used Time on Task as comparison to a previous design to show we have improved performance. So we have two sets of times to show relationship. However, it has been suggested that we can use it as a metric on its own with no comparison. That if the time by itself seems quick, this still shows efficiency in the design. Can Time on Task be used this way?

4 Answers 4

1

Pair task completion time with another metric or metrics.

Not all users, under the best circumstances, will complete a task "quickly". People have different motor abilities, reading comprehension skills, and other factors that will affect task completion times. You likely won't be able to gauge how "efficient" a design is from one round of task completion timing. To your point, it's better used as a benchmark for later comparisons.

But, if you can only do one round right now and need to know if the design is "efficient", try pairing it with other metrics. Task success rate is an obvious one - completing the task quickly but incorrectly is a problem to watch out for.

You also might give the user a quick in-study survey, like "On a scale of 1 to 10 with 1 being most difficult and 10 being easiest, how easy was it for you to complete that task?" Zoom and other platforms support in-app surveys, so it's pretty easy to capture this data on a user research call (you can also ask verbally, and have users tell you why they gave the answer they gave). Users might complete a task efficiently, but still feel that it wasn't easy, and that's good information to have.

1

I think @Izquierdo's answer covered it and is probably the right one. So I will only focus on the decision to do this, and why you were (perhaps) told that "Time on Task" is a metric in its own right.

"Time on Task" is a metric in itself, that goes without saying. But the question is how useful it's as an isolated metric.

And the answer is: it depends.

If the speed at which a task is completed is an important UX feature (e.g., to save a life or avoid a risk), then "Time on Task" is a very important metric, even in isolation.

However, if the speed at which a task is completed isn't important or it's attributable to the user's own pace, then it's an unimportant metric, whether it's isolated or in conjunction with something else. Example: If the task is to create a post using a builder, or use an LMS, or some other task that requires a more meditative cognitive process, then the only useful metric for speed is whether they can easily figure out how to complete the task. How quickly users complete these types of tasks isn't only irrelevant, but will likely give you skewed data.

In short: whoever told you this is correct, as long as you consider why to use this metric. Remember not all methods and methodologies apply to all cases.

0

Time on the task itself might not be able to give you much feedback. "if the time by itself seems quick" sounds very generic and abstract and not a very user-centric approach.

You will still be able to know how much time each task needs, and sort them by complexity to know which one could be improved or broken into sub-tasks. So personally I would end up comparing times on different tasks (not necessarily right) if comparing older versions was not an option.

0

Your task-based testing should always include some sort of user interview.

Including some interviewing with your tasks allows you to understand your target users' expectations before using the product and their reactions after using the product.

You could as the user questions about how long they would expect to spend doing 'X'. Then you could measure how long it actually takes to do 'X'. Finally, you can ask them how long they felt it took to do 'X'.

You are also partially correct in your assumption that monitoring the time it takes to complete a task has value in itself although, I would be more focused on making sure it didn't take too long than looking to see how quickly it could be done.

These three measurements taken together give you an idea of the users' understanding of the task before they complete it, their expectations of time investment, and their emotional experience of completing the task (shorter estimations for spent time relate to satisfaction in completing the task: "time flies when you're having fun")

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.